Customer Satisfaction Index (CSI)
January 2014
• Educating Professionals • Creating and Applying Knowledge • Engaging our Communities
CSI is a theoretically robust weighted
Introduction satisfaction measure for benchmarking and
tracking customer satisfaction over time
Identify the most important
services you provide to your
customers
Measure how satisfied
customers currently are with
each of these services
Perform ‘what if’ scenarios
based on hypothesised
improvements to service levels
CSI uses existing customer satisfaction
CSI Overview drivers / attributes and applies a
hierarchical approach to measure the
relationships between the items
The index is a combination of three
components:
1) The driver component measures the
importance of each service driver to
overall satisfaction
2) (optional) The attribute component
measures the importance of each
service driver attribute within a service
driver
3) (optional) The operational component
allows us to measure satisfaction as a
function of operational measures. This
function links changes in operational
performance to changes in satisfaction
with the service driver attributes, which
in turn feed into overall satisfaction
Developing i4C has considerable experience with what
we call a “Report Card” approach to
the CSI designing trade-off experiments
• The tasks are based on
comparison of a respondent’s
current satisfaction rating profile
and hypothetical (new)
satisfaction rating profiles
• The respondent is simply asked
whether their profile or the new
profile would be more satisfactory
to them.
• This forces the respondent to
consider how much each service
driver / attribute matters to them
and their choices reveal their
“weights”
Outputs - The dashboard is custom tailored to the
project outcomes and allows stakeholders
Dashboard to perform ‘what if’ scenario analysis
The final product is a dashboard
which stakeholders can use to
quickly understand the initiatives
that will provide the greatest
improvement in overall customer
satisfaction, including breakdowns
for socio-demographic segments
and other customer groups
CSI addresses many of the weaknesses
Benefits associated with standard customer
satisfaction measurement
1. Uses existing customer • No changes to current customer satisfaction survey measures
satisfaction measures • Can use previously collected customer satisfaction data in CSI
2. Choice based • Appropriate trade-off technique that derives weights from choices
• Statistical methods used for modelling are based on underlying
measurement behavioural theory
• Carefully constructed experiment allows for independent
measurement of the relationships between the drivers / attributes
and overall satisfaction
3. Accounts for problems • Does not treat each scale item as equal or linear - e.g., a score of 7
(very satisfied) is not treated the same as a score of 6 (satisfied)
with scales
• Weighted results take into account the importance of each attribute
4. Is a weighted index • Weighted results allow for relationships between the attributes,
they do not assume each attribute is independent
CSI addresses many of the weaknesses
Benefits associated with standard customer
satisfaction measurement
5. Index for each customer • Weighted index can be calculated for every individual customer
• Distributions can be viewed and used for segmentation
6. Links operations to • Operations can be linked to the scale so management are aware of
what it would take (operationally) to change satisfaction scores
satisfaction
• CSI is not a static measure
7. Decision Support System • Management can use the DSS to perform ‘what if’ scenarios
